Hall coefficient of cubic metals

A systematic study of the low-field Hall coefficient ${\mathit{R}}_{\mathit{H}}$ of 23 cubic metals is done using tabulated Slater-Koster parameters for the band structure and a tetrahedron method to calculate the Fermi-surface integrals. In the approximation of an isotropic relaxation time, the Hall coefficient depends only on the Fermi-surface topology. The effect of the band structure is reflected in deviations of ${\mathit{R}}_{\mathit{H}}$ from the free-electron values -1/ne. In favorable cases, our calculations agree to ≊±10% with experimental data. Some of the deviations can be traced back to inaccurate band structures, e.g., in Cs or to rapidly changing ${\mathit{R}}_{\mathit{H}}$ with ${\mathrm{\varepsilon }}_{\mathit{F}}$ leading to problems with convergence, e.g., in Pt. Discrepancies with experiment show the need for an anisotropic relaxation time, especially in Pd where small regions of high curvature dominate the Hall coefficient and in Al where two bands cross the Fermi surface. The similarity in the band structures of Rh, Pd, and Ag encourages the use of a rigid-band model, which agrees qualitatively with the experimental Hall coefficients for the alloys Rh-Pd and Pd-Ag.